Quasi‐Static Single‐Component Hybrid Simulation of a Composite Structure with Multi‐Axis Control

This paper presents a quasi‐static hybrid simulation performed on a single component structure. Hybrid simulation is a substructural technique, where a structure is divided into two sections: a numerical section of the main structure and a physical experiment of the remainder. In previous cases, hybrid simulation has typically been applied to structures with a simple connection between the numerical model and physical test, e.g. civil engineering structures. In this paper, the method is applied to a composite structure, where the boundary is more complex i.e. 3 degrees of freedom. In order to evaluate the validity of the method, the results are compared to a test of the emulated structure – referred to here as the reference test. It was found that the error introduced by compliance in the load train was significant. Digital image correlation was for this reason implemented in the hybrid simulation communication loop to compensate for this source of error. Furthermore, the accuracy of the hybrid simulation was improved by compensating for communication delay. The test showed high correspondence between the hybrid simulation and the reference test in terms of overall deflection as well as displacements and rotation in the shared boundary.     

Flow induced particle migration in fresh concrete: Theoretical frame,numerical simulations and experimental results on model fluids

In this paper, we describe and compare the various physical phenomena which potentially lead to flow induced particle migration in concrete. We show that, in the case of industrial casting of concrete, gravity induced particle migration dominates all other potential sources of heterogeneities induced by flow. We then show, from comparisons between experiments using model materials, dimensional analysis and numerical simulations, that, from a quantitative point of view, the viscous drag force, which prevents particles from migrating during a casting process, shall neither be computed from the apparent viscosity nor from the plastic viscosity of the suspending phase but from its tangential viscosity. Finally, the transfer of this type of numerical prediction tool to real concrete is discussed.     

Transversely compressed bonded joints

The load capacity of bonded joints can be increased if transverse pressure is applied at the interface. The transverse pressure is assumed to introduce a Coulomb-friction contribution to the cohesive law for the interface. Response and load capacity for a bonded single-lap joint was derived using non-linear fracture mechanics. The results indicated a good correlation between theory and tests. Furthermore, the model is suggested as theoretical base for determining load capacity of bonded anchorages with transverse pressure, in externally reinforced concrete structures.     

The design of an instrumented rebar for assessment of corrosion in cracked reinforced concrete

An instrumented rebar is presented which was designed to have a realistic mechanical performance and to provide location dependent measurements to assess the environment with regards to reinforcement corrosion. The instrumented rebar was constructed from a hollowed 10 mm nominal diameter standard rebar with 17 electronically isolated corrosion sensors. Instrumented and standard rebars were cast into concrete beams and bending cracks were induced and held open using steel frames. Epoxy impregnation was used to assess and compare cracks in the concrete around the instrumented and standard rebar. As bending-induced cracks reached the reinforcement, slip and separation occurred along the concrete–reinforcement interface. Cracks in the concrete surrounding standard and instrumented rebars are largely similar in appearance; however, sensors protruding from the instrumented rebar reduced the separation between the steel and concrete. Cracked beams with cast-in instrumented and standard rebars were ponded with a 10% chloride solution and the open circuit corrosion potential (OCP) of the 17 sensors was measured for up to 62 days. Measurements from the individual sensors indicate when and where active corrosion may be thermodynamically favored based upon the local environmental conditions. Results indicated the length along the instrumented rebar where active corrosion was thermodynamically favored increased with exposure time due to the increased aggressivity of the local environmental conditions.     

Simulation of residual stresses at holes in tempered glass: a parametric study

This work presents a full 3D numerical study of the residual stresses in tempered (toughened) glass near holes using Narayanaswamy’s model for the tempering process. It is the objective of the paper to elucidate the influence on the minimal residual compressive stresses at holes from variations in: the far-field stress, plate thickness, hole diameter and the interaction between holes and edges and corners. The work presents novel results for the sensitivity of the residual stresses to geometric features and provides a design tool for estimating residual stresses at holes for different geometries. An example of how to extrapolate the results in terms of far-field stresses is given.     

Failure of fibre-reinforced composites by pull-out fracture

A simple model is proposed to predict the ultimate tensile strength of fibre-reinforced composites when the failure is governed by fibre debonding. The theoretical analysis is based on the concept of fracture mechanics where the debonded zone is considered as an interfacial crack. The analysis is first applied to the classical pull-out test in order to determine the specific work of interfacial cracking. Using this value, the uniaxial tensile strength of the composites can be predicted from an approximate, closed-form equation proposed here. The theoretically predicted results seem to compare favourably with experimental values for fibrere-inforced cement based composite.     

Ultrasonic Imaging of Porosity Variations Produced During Sinteriing

A silicon carbide disk was sintered from 2090° to 2190°C in 25°C steps. After each sintering step, the disk was examined using a precision acoustic scanning system to determine acoustic attenuation and velocity. The bulk density was found to vary non-monotonically with sintering temperature. The density varied as much as 10% from its value at 2090°C during the sintering process. Local density fluctuations occurred in an organized and history-dependent way. These local density fluctuations varied up to ±7% of the bulk density and were made visible by acoustic attenuation and velocity mapping.